Elimination of hole injection barriers by conducting polymer anodes in polyfluorene light-emitting diodes

We have measured the built-in potential of a series of polyfluorene-containing light-emitting diodes (LEDs) by electroabsorption spectroscopy. When the anode incorporates a layer of poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate) (PEDOT:PSS), the inferred anode work function is equal to the ionization potential of the emissive polyfluorene. This result requires that the Fermi level of PEDOT:PSS is pinned to the highest occupied molecular orbital of the emissive polymer. Pinning of the anode Fermi level eliminates the injection barrier to holes and explains the low turn-on bias of polymer LEDs that incorporate PEDOT:PSS. We attribute this phenomenon to electron trapping at the interface between PEDOT:PSS and the emissive layer.

Figure 1

Illustration of the relationship between the built-in potential $V_{\mathrm{BI}}$, the electrode work functions, and the molecular orbitals of an organic semiconductor. (a) The energy levels when no Fermi level pinning is expected. (b) The energy levels of a completed device structure with no Fermi level pinning. (c) The energy levels when Fermi level pinning of both electrodes is expected. (d) The energy levels of a completed device structure with the Fermi level pinned at both electrodes. $\Delta$ represents the energy shift of the vacuum energy ${ϵ}_{\mathrm{vac}}$ due to interface dipoles at both electrodes.

Figure 2

Absorption (symbols) and fluorescence (line) spectra of (a) PFO, (b) F8BT, and (c) the red emitter. The repeat units of PFO and F8BT are inset.

Figure 3

$1\mathrm{\Omega }$ EA spectra of device B (open symbols) and device D (filled symbols), measured under $2\mathrm{V}$ reverse bias and $0.2\mathrm{V}$ modulation amplitude. Inset: the transmission spectra of device B (open symbols) and device D (filled symbols).

Figure 4

V scans of the $1\mathrm{\Omega }$ EA signal of 5BTF8 LEDs with ITO or PEDOT:PSS anodes and barium or aluminum cathodes, measured at $\lambda =414\mathrm{nm}$.

Figure 5

$1\mathrm{\Omega }$ EA spectrum of a PFO LED measured with a dc bias of $2\mathrm{V}$ and a sinusoidal amplitude of $0.35\mathrm{V}\mathrm{r.m.s}$. Inset: V scans of the $1\mathrm{\Omega }$ EA signal, measured at $\lambda =414\mathrm{nm}$, of LEDs with ITO (closed circles) and PEDOT:PSS (open circles) anodes.

Figure 6

V scans of the $1\mathrm{\Omega }$ EA signal of a red-doped F8BT LED, measured at $\lambda =500\mathrm{nm}$. Inset: $1\mathrm{\Omega }$ EA spectrum of a red-doped F8BT LED measured with a dc bias of $-2\mathrm{V}$. The sinusoidal amplitude was $0.2\mathrm{V}\mathrm{r.m.s.}$ for all data.

Figure 7

Illustration of the relationship between the built-in potential (filled symbols), turn-on bias (open symbols), and optical gap of the emitter for polymer LEDs using a low work-function cathode and a PEDOT:PSS anode.